SpaceX Raptor: engines for super-rockets (thread 2)

[malenfant]

All that plumbing and turbo-machinery. All those valves, actuators and sensors... It's going to be Awesome ;-)

Maybe closed-cycle gas-gas at the former scale was just too ambitious.

Complexity is largely a matter of the number of unique elements so more of the same (engines) might not be an intractable problem vs a very large engine development.

[DocM]

One way to greatly reduce the need for gimbal actuators and their plumbing & sensors is not.to use them on most of the engines. Just steer with the outer ring of engines and the center (primary landing) engine. The others can be fixed.

They could also reduce the turbomachinery parts by using one pump to drive 2-4 thrust chambers like the RD-170 family, but 3D printed pumps have massively reduced part counts (already been done) so it may not be necessary.

[chrisj1968]

1,000,000lbf? are you kidding? that would literally crack any current launchpads.

 

what would the G force be at lift off? former g's were something like 3+ G's. astronauts would have to probably be able to handle double that. insane

[DocM]

1,000,000 lbf is nothing. Easy. And remember that the engines are throttleable.

Flight G's for satellites run about 6 G's, and human ratings limit them to 3.0 to 3.5 G's. SpaceX launchers are build to human ratings and can fly either flight plan.

Falcon 9 already delivers 1,300,000 lbf, and their LC-40 pad can handle over 5,000,000 lbf. It was originally built for the Titan IV. A beast.

Their LC-39A pad is good to 12,500,000 lbf, which was built for Saturn V and used for the Shuttle. It'll launch both Falcon 9 and Falcon Heavy (4,500,000 lbf.)

And no matter how many engines it uses, BFR will need an output over 15,550,000 lbf and a new pad that can take over 18,000,000 lbf to provide margins. That and a very large safety zone for acoustic reasons. A real building rattler for about 10 miles.

[PaulRocket]

I know this is highly speculative, but what do you think or what estimates are there, to how much a reusable BFR launch to LEO cost? We might find out more later this year, but its still fun to speculate. If they want a ticket price of $500,000 per person and want to send 100 people that would equal a launch price of at least $50,000,000. Preferably even lower, since there is a lot more cost involved if you want to move to Mars then just getting there...

[DocM]

No clue, but we know SpaceX is talking 100 landed tonnes of cargo/crew per trip. That implies a total vehicle mass of at least twice that.

Something else broke recently about Raptor: they've decided to make it smaller.

They only want to design one engine for the launch phase, transits, Mars landing, and Earth return launch. A very large engine worked for all but two of these: landing on Mars and Earth return.

For the Mars legs you need a thrust to weight ratio (T/W) ranging from just below 1:1 to descend to just above 1:1 to launch back to Earth.

The 840 tonne thrust engine would deliver too much thrust to land at 100% throttle, and it's very $$ to get it to throttle down to a single digit throttle percentage for Mars landings or the Earth return launch.

The bottom line is Raptor will be scaled back to 500,000 lbf. They'll just use more of them for launch, perhaps 25 to 30 like Falcon Heavy but a single core, and at the Mars landing they only need one engine (with a spare or two) that can throttle to a more normal level.

Essentially, it appears SpaceX is going to build what is essentially an N2.

[dstarr]

"And using more engines, say 25-30, gives you an even more robust engine-out capability than Falcon 9 has - and that's pretty damned robust already.

Losing one Merlin 1D costs you 11% of your thrust, which can be made up by throttling up the remaining 8 and/or having them burn longer.

If you had a 30 engine BFR lose an engine it only costs you 3.33%. A hiccup."

 

Here are a couple of options for engine arrangements for the first stage of the BFR. The thrust values assume 500,000 lbf for each Raptor engine. The first has 23 Raptor engines and the second has 37 Raptor engines. Please pardon the crudeness of these drawings.

 

       

[DocM]

I'm thinking 27 engines, a number which they'll have experience with on Falcon Heavy.

Musk has previously said BFR would have ~15 mlbf of thrust. 15 mlbf divided by 27 gives you 555,556 lbf per engine.

[dstarr]

Their experience (soon to be) with the Falcon Heavy is with three Falcon 9 cores strapped together. In a recent interview, Elon Musk nixxed the idea of the BFR being three cores strapped together, in favor of a single "core" with a lot of engines in the one core. I'm under the impression that was due to aerodynamic force loads during the rocket ascent. That doesn't mean they can't make a single core with 27 engines, but they wouldn't be able to apply their experience with the Falcon Heavy to the BFR.

[DocM]

They could apply that part which involves starting up and running 27 engines simultaneously.

[DocM]

Raptor news coming tomorrow from the GPU Technology Conference in San Jose Cal.

https://registration.gputechconf.com/form/session-listing&doSearch=true&queryInput=&topic_selector=Supercomputing&type_selector=none

S5398 - GPUs to Mars: Full-Scale Simulation of SpaceX's Mars Rocket Engine

Day: Tuesday, 03/17

Time: 15:00 - 15:50

Location: Room LL21A

Stephen Jones Lead Software Engineer, SpaceX

Adam Lichtl Director of Research, SpaceX

SpaceX is designing a new, methane-fueled engine powerful enough to lift the equipment and personnel needed to colonize Mars. A vital aspect of this effort involves the creation of a multi-physics code to accurately model a running rocket engine. The scale and complexity of turbulent non-premixed combustion has so far made it impractical to simulate, even on today's largest supercomputers. We present a novel approach using wavelets on GPUs, capable of capturing physics down to the finest turbulent scales.

Level: All

Type: Talk

Tags: Manufacturing; Computational Physics; Supercomputing; Developer - Algorithms

[BetaguyGZT]

Those simulations are absolutely critical to the development process moving forward. Glad to see that they are keeping things open, too. That helps everyone.

 

Good on them!

[DocM]

@jeff_foust

Shotwell: vehicle architecture that will use the Raptor (large LOX/CH4) engine still in work, so engine performance TBD. #satshow<br />

[dstarr]

Here's the much desired 27 engine BFR 1st Stage arrangement. The force ratings is based on 500,000 lbf each engine. It really resembles the Soviet N1 moon rocket more than anything else.

 

 

[DocM]

A better 27-Raptor arrangement is 1 center engine surrounded by circular rings of 11 and 15 engines.

[dstarr]

I think that the proposed 1-11-15 arrangement of engine rings would be problematic because the failure of 1 engine in either the 11 or 15 engine rings would result in there being no available symmetrical engine in the ring to be able to shutdown to balance out the thrust. I think the rings of engines should be an even number to retain the best engine out capability.

[DocM]

1-10-16 or 1-8-18 solves that.

[dstarr]

1-10-16 or 1-8-18 solves that.

Of course, there are many such combinations that may be better than the 27 engine arrangement that I posted. The combinations I posted were always only meant as beginnings for this discussion we are engaged in. The 1-8-18 arrangement sounds good to me, but what is ultimately selected by SpaceX will be based on a lot more things than what we are currently privy to. Such as the physical size of all the related piping and turbines involved.

[BetaguyGZT]

And we can bet they'll be efficient buggers, too. If there's one thing to be counted on, it's that they'll improve the power-to-fuel efficiency.

 

Oh, and it'll be sodding LOUD when it launches.

[DocM]

Engine efficiency is rated in terms if total impulse, or Isp, given in seconds. To make all things equal the vacuum Isp is used for comparison.

Today's RP-1 first stage engines deliver a vacuum Isp of 338s for RD-180 and 310s for Merlin 1D (not the M1D+), the difference being their combustion cycles; RD-180's staged combustion vs. Merlin 1D's gas generator.

The methane SpaceX Raptor spec is for an Isp of 385s. The Isp for the methane Blue Origin BE-4 should be less than Raptor, again a cycle difference - Raptor's full flow staged combustion vs BE-4's standard staged combustion cycle.

Raptor's full flow staged combustion cycle has a lower pumping pressure because it's pushing gases instead of liquids, but a higher combustion chamber pressure (and thrust) due to more efficient combustion. Because of the lower pumping load full flow is generally considered more reusable.

[BetaguyGZT]

Thanks, Doc. Always the font of knowledge when called upon.

 

I think the Methane systems have great efficiency potential compared to the current-generation engines, that's why I'm particularly interested in that aspect of their performance. Evolutions of the technology should be able to demonstrate the efficiency gains that can be achieved, however small they might be.

 

With the development of better tools (3D Printing, Software, Physics Modeling, etc), Engineers and Scientists can really push the technology more quickly than ever without needing to build large prototypes unless they know they have something that will work. That saves private companies a lot of money, time, and resources.

 

It's a great time to be in the field of Rocket Science. I guess that's what I'm saying. 

[DocM]

Space is already flying printed cryogenic LOX valves on Falcon 9, flying the first one ~Q3 of last year. They've also said some Raptor parts being tested at NASA Stennis are printed, including the nozzles.

AeroJet Rocketdyne tested printed turbopump parts and found them very acceptable, and their AR-1 engine will have printed parts.

SpaceX has already qualified the printed SuperDraco engine that'll be soon be flown in the abort tests of Dragon 2.

Metals: about anything from aluminum alloys to superalloys like Inconel.

Other advantage is a massive reduction in parts count, in some cases from >100 to <10, and the ability to print shapes that are impossible on CNC etc.

Printing rocket parts and engines time is definitely here.

[DocM]

Pensacola Today (April 12 2015),

http://tinyurl.com/k4euen5

>

To say that commercial space company SpaceX is ambitious is a huge understatement. Its space vehicles supply the International Space Station and will one day transport astronauts. It

[dstarr]

When they assemble the BFR, would it not be a good idea to have the center engine be a different Raptor based engine with significantly less thrust that is also throttleable? I'm thinking it would give the landing program more options when trying to recover a BFR first stage with a vertical landing. What do you think?  Dan

[DocM]

All indications are that Raptor will be deeply throttleable, so a special landing Raptor shouldn't be necessary. A modified version of the Falcon 9 landing software should be applicable.

Let's be clear: Raptor may be the most advanced liquid rocket engine in the world when it flies. Russia and the US have tested Full Flow Staged Combustion engine components, but have never flown one. This will be a first, and parts of it are being 3D printed.